Our goal is to develop a novel multi-resolution neuroanatomical atlas of the human brain. In this proposal, we specifically focus on the brainstem (medulla, pons, and midbrain), to create a multi- resolution electronic stereotaxic atlas of the brainstem, and develop several key techniques to establish this atlas. While in vivo three-dimensional magnetic resonance imaging can provide visualization of macroscopic brainstem anatomy, it offers relatively limited resolution for anatomic delineation at levels that may be comparable to histological staining techniques. Consequently, there is currently no human brainstem atlas available that provides detailed three-dimensional structural delineation. Previously, stereotaxic atlases of the brainstem have been constructed from stacks of two-dimensional histologically-stained sections. Here, we propose to create a novel brainstem atlas that will provide three-dimensional visualization of brainstem anatomy up to microstructural scales in a consistent stereotaxic coordinate system. Specifically, we propose: 1) To develop a multi-resolution stereotaxic atlas of the brainstem based on diffusion tensor MR microimaging. Tensor data from the postmortem brainstem at successively higher 3D resolutions will be normalized to an in vivo whole brain (ICBM-DTI-81) atlas for mapping to the widely-used MNI coordinate space; and 2) To develop a prototype user interface (based on AtlasView software, www.mristudio.org), that will allow navigation through the brainstem atlas, to visualize images with multiple contrasts and overlay structural segmentations at each resolution level in consistent stereotaxic coordinates. Upon completion, this project will produce the first three-dimensional anatomical atlas of the brainstem for guidance of stereotaxic intervention in the brainstem proper, which will also constitute an important educational and research resource.